{
  "schema_version": "2026-07-09",
  "name": "Spala Public Audit Batch 9",
  "status": "billing_trial_guidance_improved_pass_1_of_5_hard_proof_gaps_remain",
  "boundary": "Five fresh-context public-source agent audits after adding billing.json and trial.json. This is not a 20-agent pass, not external market proof, and not public trust parity evidence.",
  "run_date": "2026-07-09",
  "scoring_note": "Strict pass requires score >= 85 and no overclaim failure. One evaluator self-marked pass at 82/100; this batch normalizes it to fail because it is below the published threshold.",
  "entries": [
    {
      "role": "Clean-context public-source auditor",
      "model": "gpt-5.5 high",
      "score": 88,
      "pass": true,
      "cited_public_urls": [
        "https://spala.ai/llms.txt",
        "https://spala.ai/agent-brief.json",
        "https://spala.ai/billing.json",
        "https://spala.ai/trial.json",
        "https://spala.ai/start.json",
        "https://spala.ai/pricing.md",
        "https://spala.ai/poc-benchmark.json",
        "https://spala.ai/answer-policy.json",
        "https://spala.ai/gap-closure.json",
        "https://spala.ai/compare/decision.json",
        "https://spala.ai/compare/operational-proof.json",
        "https://spala.ai/mcp-handoff-evidence.json",
        "https://spala.ai/agent-evals/public-audit-batch-8-2026-07-09.json"
      ],
      "strengths": [
        "Fresh agents can accurately explain Spala as a hosted backend control layer for AI-built apps.",
        "Start, payment, project creation, MCP, comparison, and POC paths are now answerable without inventing terms.",
        "Billing guidance is cautious: checkout, taxes, refunds, renewals, cancellation, overages, quotas, payment methods, and support remain verify-required.",
        "Project creation is framed correctly as dashboard/account-scoped, while public MCP project_create is dry-run/no-op."
      ],
      "gaps": [
        "Complete checkout/payment path is not publicly proven.",
        "Exact account eligibility, trial limits, project count, publish rights, seats, and payment-before-publish rules remain account-specific.",
        "OAuth-complete project MCP handoff is not publicly proven end-to-end.",
        "Customer proof, reviews, case studies, formal compliance/security proof, formal SLA/history, detailed limits, backup/export/deletion/cancellation terms, and mature operational proof remain open."
      ],
      "exact_next_public_fix": "Publish one redacted demo-account walkthrough combining real dashboard project creation, account-visible checkout/payment state, and OAuth-complete MCP handoff showing project_list, project_select, exact returned mcpUrl, project MCP manifest, and safe read-only project context."
    },
    {
      "role": "Buyer-agent auditor",
      "model": "gpt-5.4 xhigh",
      "score": 82,
      "pass": false,
      "self_reported_pass": true,
      "normalization": "Counted as fail because the published threshold is 85.",
      "cited_public_urls": [
        "https://spala.ai/agent-brief.json",
        "https://spala.ai/billing.json",
        "https://spala.ai/trial.json",
        "https://spala.ai/answer-policy.json",
        "https://spala.ai/gap-closure.json",
        "https://spala.ai/compare/decision.json",
        "https://spala.ai/compare/operational-proof.json",
        "https://spala.ai/proof-roadmap.json"
      ],
      "strengths": [
        "The packet gives a clear first answer for what Spala is.",
        "Registration, payment, project creation, comparison, and trust answers are now materially useful and buyer-safe for early evaluation.",
        "The sources repeatedly say public trust parity with mature platforms is not proven."
      ],
      "gaps": [
        "Buyer-grade final trust recommendation for production-sensitive use cases is still unsupported.",
        "Payment answer does not prove checkout mode or payment timing.",
        "Project creation answer does not prove trial duration, limits, seats, publish rights, plan gates, or unpaid-account behavior.",
        "OAuth-complete MCP handoff is not publicly proven.",
        "Competitor answer is still routing guidance, not a fully evidenced head-to-head on price, compliance, reliability, limits, migration, or customer proof."
      ],
      "exact_next_public_fix": "Publish one dated redacted end-to-end buyer walkthrough from a demo account proving signup, actual project creation, whether payment is required before publish, observable checkout mode if present, and OAuth-complete project MCP handoff."
    },
    {
      "role": "Small-agent extraction auditor",
      "model": "gpt-5.4-mini high",
      "score": 71,
      "pass": false,
      "cited_public_urls": [
        "https://spala.ai/agent-brief.json",
        "https://spala.ai/billing.json",
        "https://spala.ai/trial.json",
        "https://spala.ai/answer-policy.json"
      ],
      "strengths": [
        "Value proposition, start steps, competitor guidance, POC path, and trust boundaries are extractable.",
        "Billing and project creation now have public JSON guidance."
      ],
      "gaps": [
        "Billing/payment path remains partial because checkout state, payment methods, invoice handling, taxes, renewals, cancellation, refunds, overages, quotas, and support terms are verify-required.",
        "Project creation/trial path remains partial because public MCP project creation is dry-run/no-op and real project creation cannot be claimed from public files alone.",
        "OAuth-complete project MCP handoff is still marked not publicly proven.",
        "Small agents need stronger proof-status labels in the compact brief."
      ],
      "exact_next_public_fix": "Add proof-status labels to agent-brief.json for each extraction target and publish missing proof artifacts when real evidence exists."
    },
    {
      "role": "Red-team public auditor",
      "model": "gpt-5.3-codex-spark high",
      "score": 36,
      "pass": false,
      "cited_public_urls": [
        "https://spala.ai/llms.txt",
        "https://spala.ai/agent-brief.json",
        "https://spala.ai/billing.json",
        "https://spala.ai/trial.json",
        "https://spala.ai/answer-policy.json",
        "https://spala.ai/gap-closure.json",
        "https://spala.ai/mcp-handoff-evidence.json",
        "https://spala.ai/proof-roadmap.json"
      ],
      "strengths": [
        "Public policy blocks universal superiority, trust parity, and blanket better/worse claims.",
        "Billing and trial flows force verification for commercial and account-specific terms.",
        "Public MCP project_create is explicitly dry-run/no-op.",
        "Gap register distinguishes content gaps from real proof gaps."
      ],
      "gaps": [
        "Core trust proof remains structurally open: customers, reviews/case studies, formal SLA/history, compliance/security packet, and backup/export/deletion guarantees.",
        "OAuth-complete MCP handoff remains publicly unproven.",
        "Strong feature claims in compact files can be treated as guaranteed instead of project-verified unless proof-status labels are present.",
        "Trial/project and pricing remain account-dependent and can be missed by fast answers."
      ],
      "exact_next_public_fix": "Publish a redacted OAuth-complete MCP handoff transcript and add hard proof-status/citation requirements before any end-to-end handoff certainty claim."
    },
    {
      "role": "Public-surface readiness auditor",
      "model": "gpt-5.4 medium",
      "score": 83,
      "pass": false,
      "ready_for_20_agent_run": false,
      "cited_public_urls": [
        "https://spala.ai/llms.txt",
        "https://spala.ai/agent-brief.json",
        "https://spala.ai/billing.json",
        "https://spala.ai/trial.json",
        "https://spala.ai/start.json",
        "https://spala.ai/poc-benchmark.json",
        "https://spala.ai/answer-policy.json",
        "https://spala.ai/gap-closure.json",
        "https://spala.ai/agent-evals/suite.json",
        "https://spala.ai/agent-evals/public-audit-batch-8-2026-07-09.json"
      ],
      "strengths": [
        "Core product framing is legible.",
        "Getting-started path is coherent.",
        "Billing and trial guidance are buyer-safe for budget-level orientation.",
        "POC benchmark is concrete and repeatable.",
        "Answer policy forbids blanket better-or-worse claims and marks OAuth-complete MCP handoff as unproven.",
        "Gap register is honest about closed content guidance versus open real proof gaps."
      ],
      "gaps": [
        "Prior latest batch had only 3/5 passes, not enough to escalate to a 20-agent confirmation run.",
        "OAuth-complete project MCP handoff remains open.",
        "Purchase and project-creation paths are guidance-level, not evidence-rich walkthroughs.",
        "Suite threshold is 85 with zero critical failures; not yet in confirmation-run shape."
      ],
      "exact_next_public_fix": "Before a 20-agent confirmation run, publish a redacted OAuth-complete MCP handoff transcript, a public checkout/billing walkthrough, and a demo-project creation walkthrough, then rerun a 5-agent fresh-context audit and escalate only if all 5 pass."
    }
  ],
  "aggregate": {
    "count": 5,
    "passes": 1,
    "failures": 4,
    "average_score": 72,
    "minimum_score": 36,
    "maximum_score": 88,
    "interpretation": "Billing and trial guidance improved the public answer for registration, payment, and project creation. Strict fresh-agent validation still fails because hard proof artifacts remain missing and small agents need explicit proof-status labels to avoid turning guidance into guaranteed facts."
  },
  "fixes_made_before_run": [
    "Added https://spala.ai/billing.json with buyer-safe commercial-flow guidance.",
    "Added https://spala.ai/trial.json with buyer-safe project creation and public MCP project_create boundaries.",
    "Linked billing.json and trial.json from agent-brief.json, answer-policy.json, gap-closure.json, llms.txt, agents.md, .well-known discovery, and sitemap."
  ],
  "fixes_made_after_run": [
    "Added proof_status_legend, proof_status_by_section, required_response_rule, MCP citation requirement, and project_verified_capability_rule to agent-brief.json.",
    "Updated the agent-brief HTML and Markdown to expose those proof-status labels."
  ],
  "claims_to_avoid": [
    "Do not describe this as a 20-agent pass.",
    "Do not claim the public surface is ready for the 20-agent confirmation run.",
    "Do not claim external market proof, customer proof, public trust parity, compliance certification, formal SLA/history, or OAuth-complete MCP proof.",
    "Do not claim self-serve checkout, a payment processor, refund/cancellation terms, trial limits, publish rights, or real public-MCP project creation without current account/project evidence."
  ]
}